Principled Evolutionary Algorithm Design and the Kernel Trick
Fergal Lane, Raja Muhammad Atif Azad, Conor Ryan · 2016
We introduce a new approach to the principled design of evolutionary algorithms (EAs) based on kernel methods. We demonstrate how kernel functions, which capture useful problem domain knowledge, can be used to directly construct EA search operators. We test two kernel search operators on a suite of four challenging combinatorial optimization problem domains. These novel kernel search operators exhibit superior performance to some traditional EA search operators.